Thermal coking processes have been developed since the 1930s to help crude oil refineries process the “bottom of the barrel.” In general, modern thermal coking processes employ high-severity, thermal decomposition (or “cracking”) to maximize the conversion of very heavy, low-value residuum feeds to lower boiling hydrocarbon products of higher value. Feedstocks for these coking processes normally consist of refinery process streams which cannot economically be further distilled, catalytically cracked, or otherwise processed to make fuel-grade blend streams. Typically, these materials are not suitable for catalytic operations because of catalyst fouling and/or deactivation by ash and metals. Common coking feedstocks include atmospheric distillation residuum, vacuum distillation residuum, catalytic cracker residual oils, hydrocracker residual oils, and residual oils from other refinery units.
There are three major types of modern coking processes currently used in crude oil refineries (and upgrading facilities) to convert the heavy crude oil fractions (or bitumen from shale oil or tar sands) into lighter hydrocarbons and petroleum coke: Delayed Coking, Fluid Coking, and Flexicoking. These thermal coking processes are familiar to those skilled in the art. In all three of these coking processes, the petroleum coke is considered a by-product that is tolerated in the interest of more complete conversion of refinery residues to lighter hydrocarbon compounds, referred to as ‘cracked liquids’ throughout this discussion. These cracked liquids range from pentanes to complex hydrocarbons with boiling ranges typically between 350 and 950.degrees. F. In all three of these coking processes, the ‘cracked liquids’ and other products move from the coking vessel to the fractionator in vapor form. The heavier cracked liquids (e.g., gas oils) are commonly used as feedstocks for further refinery processing (e.g., Fluid Catalytic Cracking Units or FCCUs) that transforms them into transportation fuel blend stocks.
Crude oil refineries have regularly increased the use of heavier crudes in their crude blends due to greater availability and lower costs. These heavier crudes have a greater proportion of the ‘bottom of the barrel” components, increasing the need for coker capacity. Thus, the coker often becomes the bottleneck of the refinery that limits refinery throughput. Also, these heavier crudes often contain higher concentrations of large, aromatic structures (e.g., asphaltenes and resins) that contain greater concentrations of sulfur, nitrogen, and heavy metals, such as vanadium and nickel. As a result, the coking reactions (or mechanisms) are substantially different and tend to produce a denser, shot (vs. sponge) coke crystalline structure (or morphology) with higher concentrations of undesirable contaminants in the pet coke and coker gas oils. Consequently, these three coking processes have evolved through the years with many improvements in their respective technologies.
Many refineries have relied on technology improvements to alleviate the coker bottleneck. Some refineries have modified their vacuum crude towers to maximize the production of vacuum gas oil (e.g., <1050 degree F.) per barrel of crude to reduce the feed (e.g., vacuum reduced crude or VRC) to the coking process and alleviate coker capacity issues. However, this is not generally sufficient and improvements in coker process technologies are often more effective. In delayed coking, technology improvements have focused on reducing cycle times, recycle rates, and/or drum pressure with or without increases in heater outlet temperatures to reduce coke production and increase coker capacity. Similar technology improvements have occurred in the other coking processes, as well.
In addition, coker feedstocks are often modified to alleviate safety issues associated with shot coke production or ‘hot spots’ or steam ‘blowouts’ in cutting coke out of the coking vessel. In many cases, decanted slurry oil, heavy cycle oil, and/or light cycle oil from the FCCU are added to the coker feed to increase sponge coke morphology (i.e., reduce shot coke production). This increase in sponge coke is usually sufficient to alleviate the safety problems associated with shot coke (e.g., roll out of drum, plugged drain pipes, etc.). Also, the increase in sponge coke may provide sufficient porosity to allow better cooling efficiency of the quench to avoid ‘hot spots’ and steam ‘blowouts’ due to local areas of coke that are not cooled sufficiently before coke cutting. However, the addition of these materials to coker feed reduces coking process capacities.
Unfortunately, many of these technology improvements have substantially decreased the quality of the resulting pet coke. Most of the technology improvements and heavier, sour crudes tend to push the pet coke from porous ‘sponge’ coke to ‘shot’ coke (both are terms of the art) with higher concentrations of undesirable impurities: Sulfur, nitrogen, vanadium, nickel, and iron. In some refineries, the shift in coke quality may require a major change in coke markets (e.g., anode to fuel grade) and dramatically decrease coke value. In other refineries, the changes in technology and associated feed changes have decreased the quality of the fuel grade coke with lower volatile matter (VM), gross heating value (GHV), and Hardgrove Grindability Index (GHI). All of these factors have made the fuel grade coke less desirable in the United States, and much of this fuel grade coke is shipped overseas, even with a coal-fired utility boiler on adjacent property. In this manner, the coke value is further decreased.
More importantly, many of these coker technology improvements have substantially reduced the quality of the gas oils that are further processed in downstream catalytic cracking units. That is, the heaviest or highest boiling components of the coker gas oils (often referred to as the ‘heavy tail’ in the art) are greatly increased in many of these refineries (particularly with heavier, sour crudes). In turn, these increased ‘heavy tail’ components cause significant reductions in the efficiencies of downstream catalytic cracking units. In many cases, these ‘heavy tail’ components are primarily polycyclic aromatic hydrocarbons (or PAHs) that have a high propensity to coke and contain much of the remaining, undesirable contaminants of sulfur, nitrogen, and metals. In downstream catalytic cracking units (e.g., FCCUs), these undesirable contaminants of the ‘heavy tail’ components may significantly increase contaminants in downstream product pools, consume capacities of refinery ammonia recovery/sulfur plants, and increase emissions of sulfur oxides and nitrous oxides from the FCCU regenerator. In addition, these problematic ‘heavy tail’ components of coker gas oils may significantly deactivate cracking catalysts by increasing coke on catalyst, poisoning of catalysts, and/or blockage or occupation of active catalyst sites. Also, the increase in coke on catalyst may require a more severe regeneration, leading to suboptimal heat balance and catalyst regeneration. Furthermore, the higher severity catalyst regeneration often increases FCCU catalyst attrition, leading to higher catalyst make-up rates, and higher particulate emissions from the FCCU. As a result, not all coker gas oil is created equal. In the past, refinery profit maximization computer models (e.g., Linear Programming Models) in many refineries assumed the same value for gas oil, regardless of quality. This tended to maximize gas oil production in the cokers, even though it caused problems and decreased efficiencies in downstream catalytic cracking units. Some refineries are starting to put vectors in their models to properly devalue these gas oils that reduce the performance of downstream process units.